1 Libguestfs is a library for accessing and modifying guest disk images.
2 Amongst the things this is good for: making batch configuration
3 changes to guests, getting disk used/free statistics (see also:
4 virt-df), migrating between virtualization systems (see also:
5 virt-p2v), performing partial backups, performing partial guest
6 clones, cloning guests and changing registry/UUID/hostname info, and
9 Libguestfs uses Linux kernel and qemu code, and can access any type of
10 guest filesystem that Linux and qemu can, including but not limited
11 to: ext2/3/4, btrfs, FAT and NTFS, LVM, many different disk partition
12 schemes, qcow, qcow2, vmdk.
14 Libguestfs provides ways to enumerate guest storage (eg. partitions,
15 LVs, what filesystem is in each LV, etc.). It can also run commands
16 in the context of the guest. Also you can access filesystems over FTP.
18 Libguestfs is a library that can be linked with C and C++ management
19 programs (or management programs written in OCaml, Perl, Python, Ruby, Java
20 or Haskell). You can also use it from shell scripts or the command line.
22 Libguestfs was written by Richard W.M. Jones (rjones@redhat.com) and
23 hacked on by lots of other people. For discussion, development,
24 patches, etc. please use the mailing list:
26 http://www.redhat.com/mailman/listinfo/libguestfs
30 ----------------------------------------------------------------------
32 http://libguestfs.org/
36 ----------------------------------------------------------------------
38 - recent QEMU >= 0.10 with vmchannel support
39 http://lists.gnu.org/archive/html/qemu-devel/2009-02/msg01042.html
47 - XDR, rpcgen (on Linux these are provided by glibc)
49 - squashfs-tools (mksquashfs only)
51 - genisoimage / mkisofs
53 - (Optional) Augeas (http://augeas.net/)
55 - perldoc (pod2man, pod2text) to generate the manual pages and
58 - (Optional) Readline to have nicer command-line editing in guestfish.
60 - (Optional) 'reged' program from chntpw to decode Windows registry
61 entries (http://home.eunet.no/~pnordahl/ntpasswd/)
63 - (Optional) xmllint to validate virt-inspector RELAX NG schema
65 - (Optional) OCaml if you want to rebuild the generated files, and
66 also to build the OCaml bindings
68 - (Optional) OCaml xml-light, only needed if you want to rebuild the
69 virt-inspector bindings from virt-inspector.rng.
71 - (Optional) local Fedora mirror
73 - (Optional) Perl if you want to build the perl bindings
75 - (Optional) Python if you want to build the python bindings
77 - (Optional) Ruby, rake if you want to build the ruby bindings
79 - (Optional) Java, JNI, jpackage-utils if you want to build the java
82 - (Optional) GHC if you want to build the Haskell bindings
84 - (Optional) Perl XML::XPath, Sys::Virt modules (for libvirt support
87 - (Optional, but highly recommended) perl-libintl for translating perl code.
89 Running ./configure will check you have all the requirements installed
94 ----------------------------------------------------------------------
96 Then make the daemon, library and root filesystem:
98 ./configure [--with-mirror=URI]
101 Use the optional --with-mirror parameter to specify the URI of a local
102 Fedora mirror. See the discussion of the MIRROR parameter in the
103 febootstrap(8) manpage.
105 Finally run the tests:
109 If everything works, you can install the library and tools by running
110 this command as root:
116 ----------------------------------------------------------------------
118 We provide packages for Fedora >= 11 in Fedora. Use those, or build
119 from our source RPMs - it's far simpler that way.
121 You can compile libguestfs on Fedora 10 but you cannot use it with the
122 version of qemu in Fedora 10. You need to compile your own qemu, see
123 section 'qemu' below.
126 RHEL / EPEL / CentOS etc
127 ----------------------------------------------------------------------
129 We provide packages in EPEL which cover RHEL/CentOS >= 5. Use those
130 or build from our source RPMs.
134 ----------------------------------------------------------------------
136 libguestfs is now built as a package in Debian by Guido Gunther and
137 the other Debian libvirt maintainers. See:
139 http://wiki.debian.org/Teams/DebianLibvirtTeam#Packages
141 You can build for Debian in two different ways, either building a
142 Fedora-based appliance using febootstrap, yum, rpm, fakeroot,
143 fakechroot (all packaged in Debian). However the recommended way is
144 to build a Debian-based appliance using debootstrap and debirf.
146 Both ways are supported by the configure script.
150 ----------------------------------------------------------------------
152 By far the most common problem is with broken or incompatible
155 Different versions of qemu have problems booting the appliance for
156 different reasons. This varies between versions of qemu, and Linux
157 distributions which add their own patches.
159 If you find a problem, you could try using your own qemu built from
160 source (qemu is very easy to build from source), with a 'qemu
161 wrapper'. Qemu wrappers are described in the guestfs(3) manpage.
165 ----------------------------------------------------------------------
167 By default the configure script will look for qemu-kvm (KVM support).
168 You will need a reasonably recent processor for this to work. KVM is
169 much faster than using plain Qemu.
171 You may also need to enable KVM support for non-root users, by following
174 http://www.linux-kvm.org/page/FAQ#How_can_I_use_kvm_with_a_non-privileged_user.3F
176 On some systems, this will work too:
180 On some systems, the chmod will not survive a reboot, and you will
181 need to make edits to the udev configuration.
185 ----------------------------------------------------------------------
187 Previous versions of libguestfs required something called "vmchannel".
188 Vmchannel is a special device given to virtual machines which allows
189 them to communicate in some way with the host, often (but not always)
190 without using a traditional network device. In reality, there is no
191 one thing called "vmchannel". This idea has been reimplemented
192 several times under the name vmchannel, and other hypervisors have
193 their own incompatible implementation(s) too.
195 In libguestfs <= 1.0.71, we required a specific vmchannel which is
196 properly known as "guestfwd" and has been upstream in qemu since here:
198 http://lists.gnu.org/archive/html/qemu-devel/2009-02/msg01042.html
200 In libguestfs >= 1.0.71 we don't require any vmchannel implementation,
201 as long as qemu has been compiled with support for SLIRP (user mode
202 networking, or "-net user"), which is almost always the case.
204 However we still offer the ability to use vmchannel, and in future we
205 may add support for other types of qemu, which is useful in a few
206 cases, specifically where qemu packagers decide to compile out support
207 for SLIRP (qemu packagers: please don't do this).
211 ----------------------------------------------------------------------
213 If you configure with --enable-supermin then we will build a supermin
214 appliance (supermin = super-minimized). This is a very specialized
215 appliance which is built on-the-fly at runtime (specifically, when you
216 call guestfs_launch).
218 The normal appliance is a self-contained Linux operating system, based
219 on the Fedora/RHEL/CentOS Linux distro. So it contains a complete
220 copy of all the libraries and programs needed, like kernel, libc,
221 bash, coreutils etc etc.
223 The supermin appliance removes the kernel and all the executable
224 libraries and programs from the appliance. That just leaves a
225 skeleton of config files and some data files, which is obviously
226 massively smaller than the normal appliance. At runtime we rebuild
227 the appliance on-the-fly from the libraries and programs on the host
228 (eg. pulling in the real /lib/libc.so, the real /bin/bash etc.)
230 Although this process of rebuilding the appliance each time sounds
231 slow, it turns out to be faster than using the prebuilt appliance.
232 (Most of the saving comes from not compressing the appliance - it
233 transpires that decompressing the appliance is the slowest part of the
234 whole boot sequence). On my machine, a new appliance can be built in
235 under a fifth of a second, and the boot time is several seconds
238 The big advantage of the supermin appliance for distributions like
239 Fedora is that it gets security fixes automatically from the host, so
240 there is no need to rebuild the whole of libguestfs for a security
241 update in some underlying library.
243 There are several DISADVANTAGES:
245 It won't work at all except in very narrow, controlled cases like the
246 Fedora packaging case. We control the dependencies of the libguestfs
247 RPM tightly to ensure that the required binaries are actually present
250 Furthermore there are certain unlikely changes in the packages on the
251 host which could break a supermin appliance, eg. an updated library
252 which depends on an additional data file.
254 Also supermin appliances are subjected to changes in the host kernel
255 which might break compatibility with qemu -- these are, of course,
256 real bugs in any case.
258 Lastly, supermin appliances really can't be moved between branches of
259 distributions (eg. built on Fedora 12 and moved to Fedora 10) because
260 they are not self-contained and they rely on certain libraries being
261 around. You shouldn't do this anyway.
263 Use supermin appliances with caution.
266 Notes on cross-architecture support
267 ----------------------------------------------------------------------
269 At the moment we basically don't support cross-architecture or
270 32-on-64. This limits what is possible for some guests. Filesystem
271 operations and FTP export will work fine, but running commands in
272 guests may not be possible.
274 To enable this requires work for cross-architecture and 32-on-64
275 support in febootstrap, fakeroot and fakechroot.
277 The daemon/ directory contains its own configure script. This is so
278 that in future we will be able to cross-compile the daemon.
282 ----------------------------------------------------------------------
284 On my machines I can usually rebuild the appliance in around 3
285 minutes. If it takes much longer for you, use a local Fedora mirror
288 To use squid to cache yum downloads, read this first:
289 https://lists.dulug.duke.edu/pipermail/yum/2006-August/009041.html
290 (In brief, because yum chooses random mirrors each time, squid doesn't
291 work very well with default yum configuration. To get around this,
292 choose a Fedora mirror which is close to you, set this with
293 './configure --with-mirror=[...]', and then proxy the whole lot
294 through squid by setting http_proxy environment variable).
296 You will also need to substantially increase the squid configuration
298 http://fedoraproject.org/wiki/Using_Mock_to_test_package_builds#Using_Squid_to_Speed_Up_Mock_package_downloads
301 Porting to other Linux distros / non-Linux
302 ----------------------------------------------------------------------
304 libguestfs itself should be fairly portable to other Linux
305 distributions. Non-Linux ports are trickier, but we will accept
306 patches if they aren't too invasive.
308 The main porting issues are with the dependencies needed to build the
309 appliance. You will need to find or port the following packages
315 - rpm-python http://www.rpm.org/
316 - yum http://yum.baseurl.org/
317 - febootstrap http://et.redhat.com/~rjones/febootstrap/
320 Copyright and license information
321 ----------------------------------------------------------------------
323 Copyright (C) 2009 Red Hat Inc.
325 The library is distributed under the LGPLv2+. The programs are
326 distributed under the GPLv2+. Please see the files COPYING and
327 COPYING.LIB for full license information.